inform worker about failed job_done if persistent db is down

Bug #1522715 reported by surajs sharma
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Gearman
New
Undecided
Unassigned

Bug Description

Hi,

Setup:
I am running gearmand with mysql as a persistent queue.

Scene:
When the job is picked by a worker and just before worker doing a job done, the mysql goes down.
Now, if worker fire a job done, gearmand will fail to remove it from mysql and as a result of this it will stay in its internal memory queue. Now the problem here is that, worker is not informed about such failure on job done. No ERROR packet is transmitted back to the worker.

Refer line: https://github.com/dm/gearmand/blob/master/libgearman-server/server.cc#L836

at this line we should add the below code -
return _server_error_packet(GEARMAN_DEFAULT_LOG_PARAM, server_con, GEARMAN_QUEUE_ERROR, gearmand_strerror(ret), strlen(gearmand_strerror(ret)));

Let me know if i am missing something.

Regards,
Suraj Sharma

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.